Will-They-Won’t-They
收藏资源简介:
Will-They-Won’t-They是一个大规模的立场检测数据集,由剑桥大学语言技术实验室创建,包含51,284条英文推文,是目前最大的同类数据集。该数据集专注于金融领域的谣言验证任务,特别是公司间的并购(M&A)操作。数据集的创建过程涉及使用特定关键词从Twitter上抓取相关推文,并由金融领域的专家进行标注。该数据集旨在为立场检测研究提供高质量、可靠的基准,特别是在金融领域,帮助模型理解和预测市场对并购的感知,从而可能解释股票价格的重新评估。
The Will-They-Won’t-They Dataset is a large-scale stance detection dataset created by the Language Technology Lab at the University of Cambridge, comprising 51,284 English tweets and standing as the largest dataset of its kind to date. This dataset focuses on rumor verification tasks within the financial domain, specifically mergers and acquisitions (M&A) transactions between corporations. Its construction process involves scraping relevant tweets from Twitter using targeted keywords, followed by manual annotation by financial domain experts. This dataset aims to provide high-quality, reliable benchmarks for stance detection research, particularly in the financial domain, enabling models to understand and forecast market perceptions of M&A activities, thereby potentially explaining stock price reevaluations.




